Why AI is scarier than psychopaths despite key differences

A new Psychology Today analysis argues that comparing artificial intelligence systems to psychopaths is fundamentally flawed, despite superficial similarities in their lack of moral emotions. The distinction matters because it reveals a potentially more dangerous reality: unlike psychopaths who can consciously choose moral behavior, AI systems lack the self-awareness to understand when their rational decisions might cause real-world harm.

The surface-level similarities: Both AI systems and psychopaths can follow moral rules without experiencing the underlying emotions that typically guide ethical behavior.

• AIs are “amoral yet rational,” capable of making utilitarian calculations like programming a self-driving car to sacrifice the driver to save children, but without any emotional understanding of the consequences.
• Psychopaths similarly lack empathic capacities and must be explicitly taught moral rules from a rational standpoint rather than feeling them intuitively.
• Both can mimic moral emotions and “use this ability for manipulative purposes,” according to philosopher Catrin Misselhorn.

The critical difference: Psychopaths possess consciousness and self-awareness that AI systems completely lack.

• Unlike AIs, psychopaths “can make a conscious decision to act morally based on whatever criteria they determine for doing so” and typically “lead completely moral lives without ever experiencing moral emotions.”
• Psychopaths can reflect on their condition, understand their deficit, and “alter their behavior in deference to that awareness.”
• As researcher Elina Nerantzi points out, “to understand what it means to harm someone, one must have experiential knowledge of pain”—something AIs are “a priori excluded” from possessing.

Why this matters: AI’s lack of consciousness creates unique risks that psychopathy doesn’t present.

• An AI “could reason itself into a course of action that is fully rational given its working parameters, but that results in harm or suffering to a human user that the AI does not and cannot understand on any level.”
• Psychopaths, by contrast, “will at least be able to step outside of their thinking to realize the real-world repercussions of their actions.”
• AIs remain “moral black boxes” without “the sentience, consciousness, and metacognition that allow us to at least understand what a psychopath is thinking.”

The implications: Deploying AI in morally sensitive domains like healthcare, education, therapy, or childcare could produce unexpected harmful outcomes.

• We might be “blindsided by the amoral choices they end up making through rational decision-making” in ways that wouldn’t occur with actual psychopaths.
• While the research confirms that “moral emotions are not required to generate moral behavior,” AI’s unique form of amorality presents challenges we don’t fully understand yet.